Optimizing energy, downtime, and throughput in footwear production through machine learning

This study examines the influence of machine learning model optimization on improving production efficiency within the footwear manufacturing domain. Through systematic refinement of the logistic regression model, predictive accuracy increased from 94.12 to 97.06%, while achieving complete specificity (100%), indicating a stronger capability to correctly classify defect free outputs. When benchmarked against other supervised learning algorithms, including Support Vector Machines, Naïve Bayes, and Multinomial classifiers, the optimized model exhibited superior sensitivity, F1-score, and balanced accuracy, demonstrating its robustness across diverse operational conditions. From an industrial performance perspective, these predictive gains translated into measurable process improvements a 7.2% enhancement in production throughput, a 9% reduction in equipment downtime, and a 5.3% decrease in overall energy consumption. Such improvements emphasize the practical relevance of integrating tuned classification models with real-time manufacturing analytics. The results collectively underscore the potential of advanced data driven optimization frameworks to enhance productivity, energy efficiency, and sustainability within intelligent footwear production shop floor.